Order Book Imbalance Signals for Crypto Market Making

Summary

This document develops order book imbalance as an alpha input for a crypto market-making strategy. It defines static and standardized imbalance, then compares related measures: volume-adjusted mid-price (VAMP), weighted-depth order book price, and a hybrid built from effective bid and ask prices and quantities. The measures differ in how they weight bid and ask depth and how the depth range is selected. The example standardizes the imbalance time series and shifts a mid-price-based fair value by that signal. It then adjusts quotes for inventory, posts limit orders on a grid, and limits exposure with a maximum position.

The backtest examples use BTC and ETH market data, modeled latency and queue fills, and a fee-and-rebate model. The document reports that a later BTC test retained a positive return per trade, but that it fell from 0.0139% to 0.0086%, including a 0.005% rebate. This result underscores sensitivity to fee structure and fill assumptions. The examples are historical and venue-specific; modeled execution, chosen parameters, and the stated rebate limit how broadly the performance can be generalized.

Key ideas

• Order book imbalance compares displayed bid depth with ask depth and can serve as a short-term pricing signal.
• VAMP cross-weights prices and quantities across opposite sides, while weighted-depth price weights each side internally.
• The example standardizes imbalance and uses it to shift fair value before applying inventory skew to quotes.
• Quote grids, position limits, latency, queue modeling, and fees all affect market-making backtests.
• The reported return per trade declined in the later test, highlighting sensitivity to rebates and execution assumptions.

Tags

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.